The recent increase in global incidence and severity of disease caused by toxigenic Clostridium difficile presents a major clinical and diagnostic problem. C. difficile infection (CDI), ranging from mild antibiotic- associated diarrhea to potentially lethal pseudomembranous colitis, consumes a steadily increasing proportion of hospital resources for diagnosis, treatment, and infection control. Recognition of the escalation of this problem, in addition to the emergence of epidemic strains capable of toxin hyper-production and increased disease severity, have increased the urgency of improving methods for diagnosis of CDI. Disease caused by C. difficile is due to production of toxins A and B by strains harboring the toxin genes, and testing stool directly for these toxins is considered most important clinically. However, each of the existing diagnostic methods have major limitations. In particular, the most widely used tests, enzyme immunoassays (EIAs) that directly detect toxins A and B in stool, suffer from poor sensitivity and high analytical limits of toxin detection. Thereis a critical need for a simple stool test that combines high analytical sensitivity with the clinical specificity of toxin detection to allow us to optimally classify those with and without disease. Th fact that disease severity has been directly correlated to toxin levels in the host suggests that the ability to quantify toxin levels in stool could also potentially allow use of the test to predit disease outcomes and to monitor response to therapy. Quanterix Corporation (Cambridge, MA) has developed an ultrasensitive and quantitative digital ELISA technology for measurement of minute amounts of protein analytes in complex clinical specimens. Digital ELISA has limits of detection (LOD) that are consistently at least 1,000-fold lower than EIAs using the same antibodies for capture and detection. Here, Dr. Pollock (BIDMC) proposes to collaborate with Quanterix to develop and clinically validate digital ELISAs for C. difficile toxins A and B in stoo specimens. The proposed assay will be the first digital ELISA developed for a bacterial infectious diseases application and the first application of the technology using stool. Our specific aims are to 1) develop digital ELISAs for C. difficile toxins A and B with limits of detection of 1 pg/mL in stool, using commercial antibodies for capture and detection of purified native toxins A and B, and 2) clinically validate both digital ELISAs using discarded clinical specimens submitted for C. difficile testing by PCR. Concurrently with this clinical validation we will do an exploratory evaluation of the correlation of quantitative toxin A and B levels with disease severity and C. difficile strain type. This work will establish a solid technological foundation for future development of a low- cost, automated platform and for large prospective clinical studies to evaluate assay utility for predicting clinical outcome and monitoring therapeutic response. Ultimately, this work has the potential to lead to a paradigm shift in the way we diagnose and manage CDI.